A PyTorch implementation of the WaveGlow: A Flow-based Generative Network for Speech Synthesis
- Install requirements:
pip install -r requirements.txt
- Download dataset:
wget http://festvox.org/cmu_arctic/cmu_arctic/packed/cmu_us_slt_arctic-0.95-release.tar.bz2
tar xf cmu_us_slt_arctic-0.95-release.tar.bz2
-
Extract features: feature extracting pipeline is the same as tacotron
-
Training with default hyperparams:
python train.py
- Synthesize from model:
python generate.py --checkpoint=/path/to/model --local_condition_file=/path/to/local_conditon
- This is not official implementation, some details are not necessarily correct.
- Work in progress.